High Performance Computing for Healthcare & Life Sciences

Accelerate the pace of healthcare and life sciences outcomes

Better outcomes and faster science on AWS

Accelerate time to insight and go beyond the limitations of on-premises infrastructure with high performance computing (HPC) on AWS. Flexible configuration and virtually unlimited scale means you can tune and scale infrastructure as your workloads dictate, not the other way around. AWS supports a wide range of security standards and compliance certifications, including PCI-DSS, HIPAA/HITECH, FedRAMP, GDPR, FIPS 140-2, and NIST 800-171 to help you satisfy core security and compliance requirements. Fuel innovations in healthcare and life sciences with AWS’ partner network and an extensive portfolio of purpose-built HPC tools and services.

hpc-hcls-discover-the-aws-advantage

Use cases

Computational chemistry and structure based drug design

AWS HPC brings instant access to virtually unlimited computing resources to accelerate structure-based drug design. Combine the rapid advancement of protein structure solutions and faster algorithms for describing and representing flexible 3D molecular models. AWS infrastructure supports improvements in speed, accuracy and scale in virtual screening, molecular dynamics, quantum mechanics and 3-D structure solution.

Amazon EC2 instances like the C5n or C6gn can optimize compute intensive applications and the G4 can help accelerate GPU-compatible applications. You can use AWS ParallelCluster to setup and configure you applications and use Elastic Fabric Adapter (EFA) for applications that require a high bandwidth, low-latency, low-jitter network connection for inter-node communications. Amazon FSx for Lustre delivers the high-performance file system needed for multiple instances to access the same dataset. Read this blog to learn how you can setup a molecular dynamics application (GROMACS) on AWS ParallelCluster and how to select the right Amazon EC2 instance to optimize performance and price-performance depending on your workload requirements.
 

hpc-hcls-geonomics

Genomics

On AWS you can build and run data applications that are retrospective, real-time or predictive to accelerate the journey from genomic data to genomic insight. Reduce data processing times from week to hours and obtain early insights into diseases such as cancer, cystic fibrosis, and Alzheimer’s. Learn how AWS customers like AstraZeneca and Fred Hutchinson are lowering time-to-results, and running their genomics workloads at scale on AWS.

Whether you are running compute-intensive, FPGA, or GPU-accelerate based applications, you can find the right compute platform with Amazon EC2. If you are running compute intensive applications, read this blog to learn how you can significantly lower costs with AWS Graviton2. You can also use AWS Batch to orchestrate batch workloads and work with the any workflow language such as Nextflow, DRAGENCromwell, and NVIDIA Clara Parabricks to run your analysis. AWS Open Data Program houses openly available datasets, with 40+ open life sciences and genomics datasets to enable frictionless collaboration, providing research and clinical communities with a single documented source of truth.

Modeling and simulation

Access almost unlimited compute and storage resources for modeling and simulation to inform clinical trial design and predict trial outcomes. With AWS, modeling and simulation workloads such as computational fluid dynamics, pharmacokinetics, pharmacodynamics, (PK, PD), clinical trial, and systems biology simulations can satisfy computational, data standardization, and interoperability requirements. AWS customer, Relay Therapeutics was able to analyze 5 billion molecular compounds in one day vs what would have taken months using their on-premises environment.

Run clinical trial simulations on-demand by deploying Amazon EC2 GPU instances like the G4 and P4d to shorten the time and reduce costs it takes to finish them. With the AWS Deep Learning AMI and Amazon Sagemaker Notebooks you can identify potential drug candidates for new drug development and research faster and more cost effectively.

hpc-hcls-imaging

Imaging

AWS brings the broadest and deepest set of instance types, scalable storage solutions, and advanced machine learning and data integration services to meet imaging workload requirements. On AWS you can elevate image analysis and facilitate long-term data retention. HPC on AWS enables workloads including basic research, data transfer, instrument integration, diagnostic medical imaging, high content screening, digital pathology, MRI, PET, X-Ray, and CT scans. Learn how AMBRA Health expanded its medical imaging platform on AWS, and how UC San Diego leveraged AWS’ HIPPA secure environment to deploy machine learning based algorithms to X-Ray images.

Take advantage of graphics-optimized G4 Amazon EC2 instances, to facilitate medical image rendering in real-time. With AWS DataSync you can quickly transfer data for storage or processing between on-premises or cloud storage solutions or remotely stream from your on-premises workstation using the NICE DCV remote streaming protocol. You can improve identification and pattern classification in medical images with the machine-learning capabilities of AWS SageMaker.

AWS Healthcare and Life Sciences Partners

AWS collaborates with some of the leading healthcare and life sciences solution providers.

Resources

AWS collaborates with some of the leading healthcare and life sciences solution providers.

"Making stuff run fast", starting with GROMACS

Running the BWA read aligner on Graviton2 

This post highlights helps genomics researchers benchmark several different configurations, rapidly and relatively inexpensively using AWS Graviton2 instances.

Learn more

Image Recognition for COVID-19 Diagnosis w/Liverpool University & Alces Flight

Using FSx for Lustre for genomics workloads 

Learn how to build and run genomics workflows with the high-performance POSIX-compliant shared file system of Amazon FSx for Lustre at scale using AWS Batch and AWS Step Functions.

Learn more

How to accelerate CryoEM Analysis with AWS ParallelCluster and FSx for Lustre

Enabling efficient patient care using Amazon AI services

Learn how to leverage AWS healthcare specific AI/ML solutions like Amazon Comprehend Medical and Amazon Transcribe to reduce costs, improve patient care and optimize the patient experience.

Learn more

Running Genomics Workflows with Nextflow

GROMACS price-performance optimizations on AWS

This blog presents a study of performance and price of running GROMACS, a popular open-source molecular dynamics application, across a variety of Amazon Elastic Compute Cloud (Amazon EC2) instance types.

Learn more